08-102

How Can Decision

Making Be Improved?
Katherine L. Milkman
Dolly Chugh
Max H. Bazerman

Copyright © 2008 by Katherine L. Milkman, Dolly Chugh, and Max H. Bazerman

Working papers are in draft form. This working paper is distributed for purposes of comment and
discussion only. It may not be reproduced without permission of the copyright holder. Copies of working
papers are available from the author.
 How Can Decision Making Be Improved?
Katherine L. Milkman
Harvard University

Dolly Chugh
New York University

Max H. Bazerman
Harvard University

The optimal moment to address the question of how to improve human

decision making has arrived. Thanks to fifty years of research by judgment
and decision making scholars, psychologists have developed a detailed picture
of the ways in which human judgment is bounded. This paper argues that the
time has come to focus attention on the search for strategies that will improve
bounded judgment because decision making errors are costly and are growing
more costly, decision makers are receptive, and academic insights are sure to
follow from research on improvement. In addition to calling for research on
improvement strategies, this paper organizes the existing literature pertaining
to improvement strategies, highlighting promising directions for future
research.

Acknowledgements: We thank Katie Shonk and Elizabeth Weiss for their assistance.
 How Can Decision Making Be Improved? 1

Daniel Kahneman, Amos Tversky, and others have clarified the specific ways in

which decision makers are likely to be biased. As a result, we can now describe how

people make decisions with astonishing detail and reliability. Furthermore, thanks to the

normative models of economic theory, we have a clear vision of how much better

decision making could be. If we all behaved optimally, costs and benefits would always

be accurately weighed, impatience would not exist, gains would never be foregone in

order to spite others, no relevant information would ever be overlooked, and moral

behavior would always be aligned with moral attitudes. Unfortunately, we have little

understanding of how to help people overcome their many biases and behave optimally.

The Big Question

We propose that the time has come to move the study of biases in judgment and

decision making beyond description and toward the development of improvement

strategies. While a few important insights about how to improve decision making have

already been identified, we argue that many others await discovery. We hope judgment

and decision-making scholars will focus their attention on the search for improvement

strategies in the coming years, seeking to answer the question: how can we improve

decision making?

Why the Question Is Important

Errors are costly: We believe the importance of this question is somewhat self-

evident: decisions shape important outcomes for individuals, families, businesses,

governments, and societies, and if we knew more about how to improve those outcomes,

individuals, families, businesses, governments, and societies would benefit. After all,

errors induced by biases in judgment lead decision makers to undersave for retirement,
 How Can Decision Making Be Improved? 2

engage in needless conflict, marry the wrong partners, accept the wrong jobs, and

wrongly invade countries. Given the massive costs that can result from suboptimal

decision making, it is critical for our field to focus increased effort on improving our

knowledge about strategies that can lead to better decisions.

Errors will get even costlier: The costs of suboptimal decision making have

grown, even since the first wave of research on decision biases began fifty years ago. As

more economies have shifted from a dependence on agriculture to a dependence on

industry, the importance of optimal decision making has increased. In a knowledge-

based economy, we propose that a knowledge worker’s primary deliverable is a good

decision. In addition, more and more people are being tasked with making decisions that

are likely to be biased – because of the presence of too much information, time pressure,

simultaneous choice, or some other constraints. Finally, as the economy becomes

increasingly global, each biased decision is likely to have implications for a broader

swath of society.

Decision makers are receptive: Because decision making research is relevant to

businesspeople, physicians, politicians, lawyers, private citizens, and many other groups

for whom failures to make optimal choices can be extremely costly, limitations

uncovered by researchers in our field are widely publicized and highlighted to students in

many different professional and undergraduate degree programs. Those who are exposed

to our research are eager to learn the practical implications of the knowledge we have

accumulated about biased decision making so they can improve their own outcomes.

However, our field primarily offers description about the biases that afflict decision

makers without insights into how errors can be eliminated or at least reduced.
 How Can Decision Making Be Improved? 3

Academic insights await: Bolstering our efforts to uncover techniques for

improving decision making is likely to deliver additional benefits to researchers

interested in the mental processes that underlie biased judgment. Through rigorous

testing of what does and what does not improve decision making, researchers are sure to

develop a better understanding of the mechanisms underlying decision making errors.

This will deepen our already rich descriptive understanding of decision making.

What Needs to be Done to Answer the Question

Assuming we accept the importance of uncovering strategies to fend off decision-

making errors, the next question is where to begin? To address this question, we

organize the scattered knowledge that judgment and decision-making scholars have

amassed over the last several decades about how to reduce biased decision making. Our

analysis of the existing literature on improvement strategies is designed to highlight the

most promising avenues for future research on cures for biased decision making.

Debiasing Intuition: Early Failures

Before discussing successful strategies for improving decision making, it is

important to note how difficult finding solutions has proved to be. In 1982, Fischhoff

reviewed the results of four strategies that had been proposed as solutions for biased

decision making: (1) offering warnings about the possibility of bias; (2) describing the

direction of a bias; (3) providing a dose of feedback; and (4) offering an extended

program of training with feedback, coaching, and other interventions designed to improve

judgment. According to Fischhoff’s findings, which have withstood 25 years of scrutiny,

the first three strategies yielded minimal success, and even intensive, personalized

feedback produced only moderate improvements in decision making (Bazerman and

 How Can Decision Making Be Improved? 4

Moore, 2008). This news was not encouraging for psychologists and economists who

hoped their research might improve people’s judgment and decision-making abilities.

System 1 and System 2

We believe that Stanovich and West’s (2000) distinction between System 1 and

System 2 cognitive functioning provides a useful framework for organizing both what

scholars have learned to date about effective strategies for improving decision making

and future efforts to uncover improvement strategies. System 1 refers to our intuitive

system, which is typically fast, automatic, effortless, implicit, and emotional. System 2

refers to reasoning that is slower, conscious, effortful, explicit, and logical.

People often lack important information regarding a decision, fail to notice

available information, face time and cost constraints, and maintain a relatively small

amount of information in their usable memory. The busier people are, the more they

have on their minds, and the more time constraints they face, the more likely they will be

to rely on System 1 thinking. Thus, the frantic pace of life is likely to lead us to rely on

System 1 thinking much of the time and to make costly errors.

An Important Question: Can We Move from System 1 to System 2?

We believe a number of promising strategies have been uncovered for

overcoming specific decision biases by shifting people from System 1 thinking to System

2 thinking.1 One successful strategy for moving toward System 2 thinking relies on

replacing intuition with formal analytic processes. For example, when data exists on past

inputs to and outcomes from a particular decision-making process, decision makers can

construct a linear model, or a formula that weights and sums the relevant predictor

1
It should be noted that many strategies designed to reduce decision biases by encouraging System 2
thinking have proven unsuccessful. For example, performance based pay, repetition, and high stakes
incentives have been shown to have little if any effect on a wide array of biases in judgment.
 How Can Decision Making Be Improved? 5

variables to reach a quantitative forecast about the outcome. Researchers have found that

linear models produce predictions that are superior to those of experts across an

impressive array of domains (Dawes, 1971). The value of linear models in hiring,

admissions, and selection decisions is highlighted by research that Moore, Swift, Sharek,

and Gino (2007) conducted on the interpretation of grades, which shows that graduate

school admissions officers are unable to account for the leniency of grading at an

applicant’s undergraduate institution when choosing between candidates from different

schools. The authors argue that it would be easy to set up a linear model to avoid this

error (for example, by including in its calculation only an applicant’s standardized GPA,

adjusted by her school’s average GPA). In general, we believe that the use of linear

models can help decision makers avoid the pitfalls of many judgment biases, yet this

method has only been tested in a small subset of the potentially relevant domains.

Another System 2 strategy involves taking an outsider’s perspective: trying to

remove oneself mentally from a specific situation or to consider the class of decisions to

which the current problem belongs (Kahnmean and Lovallo, 1993). Taking an outsider’s

perspective has been shown to reduce decision makers’ overconfidence about their

knowledge (Gigerenzer, Hoffrage, & Kleinbölting, 1991), the time it would take them to

complete a task (Kahneman & Lovallo, 1993), and their odds of entrepreneurial success

(Cooper, Woo, and Dunkelberg, 1988). Decision makers may also be able to improve

their judgments by asking a genuine outsider for his or her view regarding a decision.

Other research on the power of shifting people toward System 2 thinking has

shown that simply encouraging people to “consider the opposite” of whatever decision

they are about to make reduces errors in judgment due to several particularly robust
 How Can Decision Making Be Improved? 6

decision biases: overconfidence, the hindsight bias, and anchoring (Larrick, 2004;

Mussweiler, Strack, & Pfeiffer, 2000). Partial debiasing of errors in judgment typically

classified as the result of “biases and heuristics” (see Tversky and Kahneman, 1974) has

also been achieved by having groups rather than individuals make decisions, training

individuals in statistical reasoning, and making people accountable for their decisions

(Larrick, 2004; Lerner & Tetlock, 1999).

One promising debiasing strategy is to undermine the cognitive mechanism that is

hypothesized to be the source of bias with a targeted cue to rely on System 2 processes

(Slovic and Fischhoff, 1977). In a study designed to reduce hindsight bias (the tendency

to exaggerate the extent to which one could have anticipated a particular outcome in

foresight), Slovic and Fischhoff developed a hypothesis about the mechanism producing

the bias. They believed that hindsight bias resulted from subjects’ failure to use their

available knowledge and powers of inference. Armed with this insight, Slovic and

Fischhoff hypothesized and found that subjects were more resistant to the bias if they

were provided with evidence contrary to the actual outcome. This result suggests that the

most fruitful directions for researchers seeking to reduce heuristics and biases may be

those predicated upon “some understanding of and hypotheses about people’s cognitive

processes” (Fischhoff, 1982) and how they might lead to a given bias. Along these lines,

another group of researchers hypothesized that overclaiming credit results from focusing

only on estimates of one’s own contributions and ignoring those of others in a group.

They found that requiring people to estimate not only their own contributions but also

those of others reduces overclaiming (Savitsky, Van Boven, Epley, and Wight, 2005).
 How Can Decision Making Be Improved? 7

Another promising stream of research that examines how System 2 thinking can

be leveraged to reduce System 1 errors has shown that analogical reasoning can be used

to reduce bounds on people’s awareness (see Bazerman and Chugh 2005 for more on

bounded awareness). Building on the work of Thompson, Gentner, and Loewenstein

(2000), both Idson, Chugh, Bereby-Meyer, Moran, Grosskopf, and Bazerman (2004) and

Moran, Ritov, and Bazerman (2008) found that individuals who were encouraged to see

and understand the common principle underlying a set of seemingly unrelated tasks

subsequently demonstrated an improved ability to discover solutions in a different task

that relied on the same underlying principle. This work is consistent with Thompson et

al.’s (2000) observation that surface details of learning opportunities often distract us

from seeing important underlying, generalizable principles. Analogical reasoning

appears to offer hope for overcoming this barrier to decision improvement.

Work on joint-versus-separate decision making also suggests that people can

move from suboptimal System 1 thinking toward improved System 2 thinking when they

consider and choose between multiple options simultaneously rather than accepting or

rejecting options separately. For example, Bazerman, White and Loewenstein (1995)

find evidence that people display more bounded self-interest (Jolls, Sunstein, and Thaler,

1998) – focusing on their outcomes relative to those of others rather than optimizing their

own outcomes – when assessing one option at a time than when considering multiple

options side by side. Bazerman, Loewenstein and White (1992) have also demonstrated

that people exhibit less willpower when they weigh choices separately rather than jointly.

The research discussed above suggests that any change in a decision’s context that

promotes cool-headed System 2 thinking has the potential to reduce common biases
 How Can Decision Making Be Improved? 8

resulting from hotheadedness, such as impulsivity and concern about relative outcomes.

Research on joint-versus-separate decision making highlights the fact that our first

impulses tend to be more emotional than logical (Moore and Loewenstein, 2004). Some

additional suggestive results in this domain include the findings that willpower is

weakened when people are placed under extreme cognitive load (Shiv and Fedorkihn,

1999) and when they are inexperienced in a choice domain (Milkman, Rogers and

Bazerman, 2008). Other research has shown that people make less impulsive, sub-

optimal decisions in many domains when they make choices further in advance of their

consequences (see Milkman, Rogers and Bazerman, in press, for a review). A question

we pose in light of this research is when and how carefully selected contextual changes

promoting increased cognition can be leveraged to reduce the effects of decision making

biases?

Another Important Question: Can We Leverage System 1 to Improve Decision Making?

Albert Einstein once said, “We can't solve problems by using the same kind of

thinking we used when we created them.” However, it is possible that the unconscious

mental system can, in fact, do just that. In recent years, a new general strategy for

improving biased decision making has been proposed that leverages our automatic

cognitive processes and turns them to our advantage (Sunstein and Thaler, 2003). Rather

than trying to change a decision maker’s thinking from System 1 to System 2, this

strategy tries to change the environment so that System 1 thinking will lead to good

results. This type of improvement strategy, which Thaler and Sunstein discuss at length

in their book Nudge (2008), calls upon those who design situations in which choices are

made (whether they be the decision makers themselves or other “choice architects”) to
 How Can Decision Making Be Improved? 9

maximize the odds that decision makers will make wise choices given known decision

biases. For example, a bias towards inaction creates a preference for default options

(Ritov and Baron, 1992). Choice architects can use this insight to improve decision

making by ensuring that the available default is the option that is likely to be best for

decision makers and/or society. Making 401k enrollment a default, for instance, has been

shown to significantly increase employees’ savings rates (Benartzi and Thaler, 2007).

There is also some suggestive evidence that leveraging System 1 thinking to

improve System 1 choices may be particularly effective in the realm of decision-making

biases that people do not like to admit or believe they are susceptible to. For instance,

many of us are susceptible to implicit racial bias but feel uncomfortable acknowledging

this fact, even to ourselves. Conscious efforts to simply “do better” on implicit bias tests

are usually futile (Nosek, Greenwald, & Banaji, 2007). However, individuals whose

mental or physical environment is shaped by the involvement of a black experimenter

rather than a white experimenter show less implicit racial bias (Lowery, Hardin, &

Sinclair, 2001; Blair, 2002). The results of this “change the environment” approach

contrast sharply with the failure of “try harder” solutions, which rely on conscious effort.

In summary, can solutions to biases that people are unwilling to acknowledge be found in

the same automatic systems that generate this class of problems?

Conclusion

People put great trust in their intuition. The past 50 years of decision-making

research challenges that trust. A key task for psychologists is to identify how and in what

situations people should try to move from intuitively compelling System 1 thinking to

more deliberative System 2 thinking and to design situations that make System 1 thinking
 How Can Decision Making Be Improved? 10

work in the decision-maker’s favor. Clearly, minor decisions do not require a complete

System 2 process or a new decision architecture. However, the more deeply we

understand the repercussions of System 1 thinking, the more deeply we desire empirically

tested strategies for reaching better decisions. Recent decades have delivered description

in abundance. This paper calls for more research on strategies for improving decisions.

References
Bazerman, M. H. & Chugh, D. (2005). Bounded awareness: Focusing failures in negotiation. In L.
Thompson (Ed.), Frontiers of Social Psychology: Negotiation. Psychological Press.

Bazerman, M. H., Loewenstein, G. F., & White, S. B. (1992). Reversals of preference in allocation
decisions: Judging an alternative versus choosing among alternatives. Administrative Science Quarterly,
37(2), 220-240.

Bazerman, M.H. & Moore, D. (2008). Judgment in Managerial Decision Making (7th ed.). Hoboken, NJ:
John Wiley & Sons, Inc.

Bazerman, M.H., White, S.B., & Loewenstein, G.F. (1995). Perceptions of Fairness in Interpersonal and
Individual Choice Situations. Current Directions in Psychological Science, 4, 39-43.

Blair, I. V. (2002). The malleability of automatic stereotypes and prejudice. Personality & Social
Psychology Review, 6(3), 242-261.

Benartzi, S., & Thaler, R. H. (2007). Heuristics and biases in retirement savings behavior. Journal of
Economic Perspectives, 21(3), 81-104.

Cooper, A. C., Woo, C. Y., & Dunkelberg, W. C. (1988). Entrepreneurs' perceived chances for success.
Journal of Business Venturing, 3(2), 97-109.

Dawes, R. M. (1971). A case study of graduate admissions: Application of three principles of human
decision making. American Psychologist, 26(2), 180-188.

Fischhoff, B. (1982). Debiasing. In D. Kahneman, P. Slovic, & A. Tversky (Eds.), Judgment Under
Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press: 422 – 444.

Idson, L. C., Chugh, D., Bereby-Meyer, Y., Moran, S., Grosskopf, B., & Bazerman, M. (2004).
Overcoming focusing failures in competitive environments. Journal of Behavioral Decision Making, 17(3),
159-172.

Johnson, Eric J. and Goldstein, D. (2003). Science. 302 (5649), 1338 – 1339.

Jolls, C., Sunstein, C. R., & Thaler, R. (1998) A behavioral approach to law and economics. Stanford Law
Review, 50, 1471-1550

Kahneman, D., & Lovallo, D. (1993). Timid choices and bold forecasts: A cognitive perspective on risk
and risk taking. Management Science, 39, 17-31.

Larrick, R. P. (2004). Debiasing. In D. J. Koehler & N. Harvey (Eds.), Blackwell Handbook of Judgment
and Decision Making. Oxford, England: Blackwell Publishers.

Lerner, J. S., & Tetlock, P. E. (1999). Accounting for the effects of accountability. Psychological Bulletin,
125(2), 255-275.
 How Can Decision Making Be Improved? 11

Lowery, B. S., Hardin, C. D., & Sinclair, S. (2001). Social influence effects on automatic racial prejudice.
Journal of Personality and Social Psychology, 81(5), 842-855.

Milkman, K.L., Rogers, T., & Bazerman, M. (2008). Highbrow films gather dust: A study of dynamic
inconsistency and online DVD rentals. HBS Working Paper 07-099.

Milkman, K.L., Rogers, T., & Bazerman, M. (in press). Harnessing our inner angels and demons: What
we have learned about want/should conflict and how that knowledge can help us reduce short-sighted
decision making. Perspectives on Psychological Science.

Moore, D., & Lowenstein, G. (2004). Self-interest, automaticity, and the psychology of conflict of interest.
Social Justice Research, 17(2), 189-202.

Moore, D. A., Swift, S. A., Sharek, Z., & Gino, F. (2007). Correspondence bias in performance evaluation:
Why grade inflation works. Tepper Working Paper 2004-E42.

Moran, S., Ritov, I., & Bazerman, M.H. (2008). Details need to be added.

Mussweiler, T., Strack, F., & Pfeiffer, T. (2000). Overcoming the inevitable anchoring effect: Considering
the opposite compensates for selective accessibility. Personality and Social Psychology Bulletin, 26, 1142-
1150.

Nosek, B. A., Greenwald, A. G., & Banaji, M. R. (2007). The Implicit Association Test at age 7: A
methodological and conceptual review. In J. A. Bargh (Ed.), Social Psychology and the Unconscious: The
Automaticity of Higher Mental Processes, New York: Psychology Press.

Ritov, I., & Baron, J. (1992). Status-quo and omission biases. Journal of Risk & Uncertainty, 5(1), 49-61.

Savitsky, K., Van Boven, L., Epley, N., & Wight, W. (2005). The unpacking effect in responsibility
allocations for group tasks. Journal of Experimental Social Psychology, 41, 447-457.

Shiv, B., & Fedorikhin, A. (1999). Heart and mind in conflict: The interplay of affect and cognition in
consumer decision making. Journal of Consumer Research, 26(3), 278-292.

Slovic, P. & Fischhoff, B. (1977). On the psychology of experimental surprises. Journal of Experimental
Psychology: Human Perception and Performance, 3, 544-551: 23, 31.

Stanovich, K. E., & West, R. F. (2000). Individual differences in reasoning: Implications for the rationality
debate. Behavioral & Brain Sciences, 23, 645-665.

Sunstein, C.R. & Thaler, R.H (2003). Libertarian paternalism is not an oxymoron. University of Chicago
Law Review, 70 (Fall), 1159-99.

Thaler, R. H., & Sunstein, C. R. (2008). Nudge. New Haven: Yale University Press.

Thompson, L., Gentner, D., & Loewenstein, J. (2000). Avoiding missed opportunities in managerial life:
Analogical training more powerful than individual case training. Organizational Behavior & Human
Decision Processes, 82(1), 60-75.

Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science,
185(4157), 1124-1131.